This paper details a novel procedure for accurately classifying lower facial expressions. A shape model is developed based on an anatomical analysis of facial expression called the Facial Action Coding System (FACS). This model analyses the movement in shape due to the formation of a specific expression.
We apply Kernel Principal Component Analysis (K.PCA) to the shapes in the training set and classify new unseen expressions by using Support Vector Machines (SVMs). We further analyse our model by attaching a probability measure to the outputs.
